SUMMARY This project aims to produce and beta-test a clinical decision support application for end- stage heart failure, titled CHRiSS: Cardiac Health Risk Stratification System. It will be designed for patients with progressive heart failure who may eventually become candidates for ventricular assist device (VAD) therapy. The software will be build upon existing machine learning and data mining technology, developed by the PI and colleagues, designed to predict 90-day mortality following VAD implantation. This Phase- 1 effort will entail development of graphic user interfaces for both patients and clinicians to incorporate personalized prognostic information into a functional decision-support utility. The inference algorithm will also be expanded to encompass adverse events, and risk of readmission. In this Phase-1 feasibility project, CHRiSS will be programmed and beta-tested at West Penn Allegheny Health System (WPAHS). This will entail semi- structured interviews to collect relevant expert knowledge, and mining the electronic medical records (EMR) to calibrate the prognostic algorithm to the institution-specific data. Successful completion of this Phase I project will lead to a multi-center trial in which the software is fully validated. Ultimate succes of this project will result in a software application that will optimize the benefit of VAD therapy to public health, while reducing cost by reducing the unacceptably high rate of adverse events and unnecessary readmissions.